Global sleep overview (3 subjects Night Owl Early Bird Sleepy Bear)

y variable defined on the left remains the same

total sleep overview

total sleep overview (by day of week)

deep sleep variation (by day of week)

_stress_awake (by day of week)

_stress_awake

Influence from Median_hr_sleep on sleep quality

Influence from median_stress_sleep on sleep quality

Correlation stress & HR

The p-value of the test is less than the significance level alpha = 0.05. We can conclude that hr sleep and stress sleep are significantly correlated with a correlation coefficient beetween 0.64–0.88.

Correlation vigorous_activity_hours & total_sleep_hours

The p-value of the test is more than the significance level alpha = 0.05. We can conclude that variable are not significantly correlated .

Correlation steps & total_sleep_hours

The p-value of the test is more than the significance level alpha = 0.05. We can conclude that variable are not significantly correlated .

**Correlation %_high_stress_30min_before_bedtime & total_sleep_hours**

The p-value of the test is more than the significance level alpha = 0.05. We can conclude that variable are not significantly correlated .

Correlation estimated_sleep_time & total_sleep_hours

The p-value of the test is less (2 participants) than the significance level alpha = 0.05. We can conclude that variable are significantly correlated .

Correlation

ML decision tree

n= 122 

node), split, n, loss, yval, (yprob)
      * denotes terminal node

 1) root 122 23 0 (0.81147541 0.18852459)  
   2) total_sleep_hours>=7.283333 84  9 0 (0.89285714 0.10714286)  
     4) light_sleep_percentage< 0.7714757 79  6 0 (0.92405063 0.07594937)  
       8) light_sleep_percentage>=0.547778 67  3 0 (0.95522388 0.04477612) *
       9) light_sleep_percentage< 0.547778 12  3 0 (0.75000000 0.25000000)  
        18) awake_percentage< 0.00330662 7  0 0 (1.00000000 0.00000000) *
        19) awake_percentage>=0.00330662 5  2 1 (0.40000000 0.60000000) *
     5) light_sleep_percentage>=0.7714757 5  2 1 (0.40000000 0.60000000) *
   3) total_sleep_hours< 7.283333 38 14 0 (0.63157895 0.36842105)  
     6) light_sleep_percentage< 0.5919124 20  4 0 (0.80000000 0.20000000) *
     7) light_sleep_percentage>=0.5919124 18  8 1 (0.44444444 0.55555556)  
      14) awake_percentage< 0.004608319 3  0 0 (1.00000000 0.00000000) *
      15) awake_percentage>=0.004608319 15  5 1 (0.33333333 0.66666667)  
        30) %_stress_awake>=0.520264 6  2 0 (0.66666667 0.33333333) *
        31) %_stress_awake< 0.520264 9  1 1 (0.11111111 0.88888889) *

ML using glm( with bad_sleep_outlier_binary ~ ., family = binomial(link = “logit”)

Call:
glm(formula = bad_sleep_outlier_binary ~ ., family = binomial(link = "logit"), 
    data = train)

Deviance Residuals: 
     Min        1Q    Median        3Q       Max  
-1.34027  -0.68323  -0.35048  -0.02378   2.90516  

Coefficients: (1 not defined because of singularities)
                                  Estimate Std. Error z value Pr(>|z|)   
(Intercept)                     -3.055e+00  5.219e+00  -0.585   0.5583   
total_sleep_hours               -5.221e-01  2.971e-01  -1.757   0.0789 . 
light_sleep_percentage           9.375e+00  5.501e+00   1.704   0.0883 . 
deep_sleep_percentage            6.664e+00  4.786e+00   1.392   0.1638   
rem_sleep_percentage                    NA         NA      NA       NA   
awake_percentage                 2.967e+00  6.103e+00   0.486   0.6268   
mean_hr_sleep                    1.396e-01  4.534e-01   0.308   0.7582   
median_hr_sleep                 -1.046e-01  4.429e-01  -0.236   0.8132   
activity_calories               -4.179e-03  4.122e-03  -1.014   0.3106   
activity_hours                   5.481e-01  7.571e-01   0.724   0.4691   
steps                            1.643e-04  1.622e-04   1.013   0.3112   
distance                        -1.222e-05  1.048e-04  -0.117   0.9072   
moderate_activity_hours         -3.414e+00  2.706e+00  -1.262   0.2070   
vigorous_activity_hours          7.689e+00  3.811e+00   2.017   0.0437 * 
floors_climbed                  -3.169e-03  1.558e-02  -0.203   0.8388   
mean_hr_awake                   -2.403e-01  1.309e-01  -1.836   0.0663 . 
`%_stress_awake`                 7.342e-01  2.265e+00   0.324   0.7459   
mean_hr_30min_before_bedtime     1.748e-01  6.492e-02   2.692   0.0071 **
`%_stress_30min_before_bedtime` -2.617e-01  1.152e+00  -0.227   0.8202   
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for binomial family taken to be 1)

    Null deviance: 129.312  on 117  degrees of freedom
Residual deviance:  94.087  on 100  degrees of freedom
AIC: 130.09

Number of Fisher Scoring iterations: 6